There was a discussion on the economist podcast about this.

1. Passenger jets often take a bit of cargo, but even if they rejiggered, it'd only be half full of cargo.

2. Many companies already hedged the price of fuel so the oil glut isn't really helping them any.

3. Demand is unidirectional: medical supplies are made in a handful of countries (ie China) but there isn't a huge demand for air freight the other direction. See also the container problem.

4. The airline network is a complex emergent system. Passengers from one international flight often connect into another domestic one. Cancelling an international flight has unpredictable knock-on effects for downstream flights.

5. Cancelling a flight can lead to losing your 'slot'. Similar to how google will kill non performing ads, airports have a system to allocate arrivals / departures to airlines that factors in usage. If you shift a passenger jet flying EU to US into doing cargo hauls, that slot is jeopardized.

> 3. Demand is unidirectional: medical supplies are made in a handful of countries (ie China) but there isn't a huge demand for air freight the other direction. See also the container problem.

This has got to change. Why is everything made in China? Anyone not coming to grips with the national security implications of not creating critical equipment in your own country (or region) is praying for doom.

> Why is everything made in China?

Part of me wonders how serious a question this is. The ridiculously cheap manufacturing costs is why anything is made in China. When the Chinese get unions to negotiate $45/hour salaries, require fair working conditions, time off, medical/dental, and on and on things might be different.

You can take advantage of the situation and develop a 100% robotic assembly line that does not require $45/hour workers so that you can manufacture critical equipment in your own country. Be the hero the world needs, not the one it deserves.

While the internet is highly variable in this question, it appears that for the cases where it is cheaper to manufacture in china, the cost of manufacturing is only about 5% cheaper.

In some other cases where all the components are already made in the states, it can be upwards of 20% cheaper to make it here, because of the shorter lead time, less complex supply chain, as well as other factors.

The cost differential of China has effectively vanished due to rising wages, the implementation of some environmental rules, and increasing strength of the RMB.

What China does have is well established closely knit supply chains, which means its a shorter lead-time to start new production of a given widget - this is something that we can fix over time, and in my opinion, this may be the biggest lasting change from this crisis. We effectively got caught with our pants down in this crisis because of this.

This is anecdotal but I have yet to find a PCB manufacturer in the US that can do it cheaper than Chinese fabs at anything less than ridiculous scale - high speed digital/RF included. I still use US manufacturers half the time because of IP, timeline, or quality concerns (its just easier to get on the phone with someone who can help) but I don't see how the other manufacturers could be within 5% of China's cost when circuit boards at anywhere from 2x to 10x more expensive here.

You probably know them, but oshpark is great! You pay by the square inch, which was really fortunate for me because I needed a (very) small pcb for a project, and they made and shipped it to me (to Germany!) for less than 2$ at the time. Though I think they since implemented a lower bound on size/price.

Also, the purple/gold finish is gorgeous.

I'm curious, what run size are you looking at?

From ten to about one million. There are major caveats though: The middle volume boards (4-5 figures) were far more complex mixed signal boards with 10-20 layers and clock speeds ranging from several hundred Mhz for RAM busses to 1 ghz for MIPI to 2-50ghz for radios. Those projects had special needs and required quite a bit more manpower during the fabrication stage for layer alignment and in the assembly stage for tuning, since they exploited some nonlinearities in different components for high precision synchronization.

The large volume boards (6-7 figures) where consumer electronics focused and extremely price sensitive so the combined savings from colocation of fab and assembly were often significantly more than any savings automation or logistics in the United States, especially accounting for risk.

Where did you pull that 5% number out from?

I visited a factory here in the UK last year (with some school students), who make plastic injection moulded parts and items. The CEO there said that yes they do manufacture a proportion of their goods in China but the reason that they still manufacture in the UK was that China was a long way away and in fact didn't turn out that much cheaper in the end so there was still money to be made by having a more responsive supply chain with more hands-on quality control. There is a lot of automation in injection moulding though.

I saw it quoted in three different blogspam articles, I basically googled and opened the first 10 results. I was having trouble finding a hard study - but people I know who work in manufacturing have quoted me "for us its cheaper to make it here" to "up to 20% more" so the 5% number didnt feel out of place.

Everyone thinks it is cheaper in China, but reality is probably more complicated. Remember 10 years ago (probably every year ;-)) when companies started planning for more international software development cause you could pay Chinese and Indian developers less? Then sometimes it didn't work out and their salaries got higher.

The manufacturing industry local skill availability (manufacturing engineers?) must have degraded here in the US.

I've investigated this question a lot and come to conclusion:

1. It's network effect and lack of regulations.

I can buy a $50 welder from china, which European or American company can sell me $50 welder delivered to my home?

2. I am in India now and to get any company to even to listen to me, they'll require several phone calls and then they'll quote me some rediculous price, no one responds to any email or text in queries. But Chinese are happy to do so.

Chinese are happy to declare lower value in invoice so that importer can lower the import duties at their end.

3. If you can arrange a logistic partner, any Chinese company will drop the goods/samples to your local wharehouse within a day or two.

4. Even illiterate or village laborers can work in Chinese factory as long as they are getting work done. Heck, they can even create new product for their employer and get paid handsomely for that. I know one Chinese guy who today owns an electronic company, he dropped out in highschool and started selling custom designed electronics. In how many countries, you can design electronic device and sell it without undergoing various checks?

5. One of my friend imports Chinese product (with made in India mark) and inserts a bolt at the destination and gets all the publicity for making in India in local newspapers. Pretty, sure it's also true for many European companies who get everything done in china.

You can start at very bottom, you need to follow any rules as long as you aren't causing big problems for everyone, you are in business.

I've spent time in China, Europe, US sourcing products for my customers. China is completely different ballgame, you need very little money to get started in Chinese business.

My team was trying very hard to source locally made goods from india, now we gave up and went all china.

Proximity to the raw materials is another important factor. Latex gloves are mainly made in Asia, because that's where most of the rubber trees are.

I tried getting into 3D printing filament business in India and result? I couldn't find raw pellet providers which are manufactured locally and had to import the pellets, after paying duties even cheap labor couldn't make me competitive as prices as some American or European or Chinese brands.

Most rubber is synthetic (including nitrile (not latex) exam gloves)

There are a huge number of rubber trees in Africa, too. Firestone is/was one of the largest employers in Liberia. The rubber plantations in Ghana are massive.

The problem is China is already building the robotic assembly lines. They have been purchasing more advanced CNC machining equipment than the rest of the world combined and the volume is increasing. Their quality should continue to increase and costs decrease even as their labor begins to organize itself.

> Part of me wonders how serious a question this is. The ridiculously cheap manufacturing costs is why anything is made in China. When the Chinese get unions to negotiate $45/hour salaries, require fair working conditions, time off, medical/dental, and on and on things might be different.

I've often wondered why import tariffs and/or local manufacturing subsidies aren't used to address this. AFAIK they're set fairly arbitrarily, based on the whims of e.g. President Trump. But imagine if they approximated the difference in cost due to the environmental and labor standards of the two countries. Then it'd be an even playing field; there'd be no advantage in importing from a country with lower standards or exporting to a country with higher standards.

A few possible reasons why not: it seems non-trivial to accurately determine how much labor went into a part (may vary over time and by manufacturer) or how much environmental damage its manufacture caused. And maybe we just don't have the appetite to pay the cost of making things to our alleged standards and would rather dump the environmental cost on China. Even local manufacturers might fight it as they don't like the precedent of considering externalities.

> I've often wondered why import tariffs and/or local manufacturing subsidies aren't used to address this.

Partially due to WTO restrictions and feigned ignorance. You're right that it's non-trivial to properly assess the P&L of outsourced labor especially when most of the externalities (ex. emissions spreading across East Asia, population health) are not immediately identifiable. Globalization ignores these externalities in favor of short term profit. To put a price tag on such effects would be to acknowledge them, which nobody will do.

Actually this is what India does and result? If I can't get my $50 welder from china because government imposed $50 import duty on it, then I am not likely to invest into learning welding. And that job is going to some person in other country where government isn't doing this.

Government should stay away from trying to control import/export and see the local labor/business adjust on its own.

In manufacturing garbage in = garbage out. Use shoddy tools, end up with shoddy quality product.

It might work for domestic demand.

But internationally? A guy in India won't care about ethics used behind the product as long as he's getting his welder unit for the price/quality ratio he's aiming for. Usually, the volumes of sales accured in this fashion will result in efficiency beyond what cheap labor/lack of regulations can get.

> I've often wondered why import tariffs and/or local manufacturing subsidies aren't used to address this.

Have you been paying attention the last couple of years? Every time Trump tried to impose tariffs in order to prop up American manufacturers he faced no end of hostility, endless explanations about why tariffs don’t work, etc.

Hopefully this whole fiasco has opened people’s eyes to the fact that we should try to bring home some amount of domestic manufacturing instead of just saying “those jobs are never coming back” like a certain former president once said.

and then when someone tries to make a union for 45 in china someone else will come up cheaper and that new factory will win all of the old ones business. and since no loyalty all the old contracts will switch over in a blink. also the expensive license will just magically "lose it's business license" when a union is tried and only one without a union will have a active license. Lots of reasons things are the way they are versus usa. are we willing to deliver on a scooter for 2 dollars an hour or sell 1-5 dollar meals from our little stall with several dishes and live like that and be happy?

My understanding (at least with the iPhone) was that assembly/test is <2% of the cost, so doubling or even tripling that component isn't that significant. It's the supply chain that also needs to be aligned.

The "raw materials" used in that cost estimate from Foxconn are things like Samsung displays made in Korea, but also Apple SoCs and Qualcomm modems made at TSMC in Taiwan, and memory, power management, RF, cameras, etc. etc. etc. made in China. The time for Foxconn to assemble those modules, but an "Assembled in the US from parts made in China" iPhone would have the same difficulties. If you add the time to build the subcomponents and their support overhead it would cost much more. Also, the iPhone has a uniquely large margin and non-recoverable engineering costs, and the semiconductor and PCB industry is necessarily highly automated. The average bit of consumer goods has a larger assembly ratio.

For what an anecdote is worth (to give a sense of the scale of the problem) I'm involved in automotive tier 2 industrial automation in the midwestern US. One of my customers who has plants in several countries has told me that a fully loaded hour (including benefits, taxes, etc) of assembly operator time in Michigan costs $28, the same hour in China costs $7.50, and the same hour in Mexico costs $4.75. It is a bit more of a pain to maintain qualified engineering and maintenance parity across these different zones, but they basically do the automation in Michigan and mirror those known-good machines and processes overseas with remote support. They're assembling, staking, polishing and inspecting plastic interior components, and their costs are roughly 15% overhead and profit, 20% shipping, 20% raw materials, 10% capital (machine) depreciation, and 35% operator time.

supply chains and competency.

the component manufacturers are all right there, which simplifies things.

plus china is really good at building stuff. you want a factory to make something? their ability to spin up production is excellent.

the pay side of it is really much less of a factor than people think. china's just better at this stuff right now.

Yeah initially it was moved for pay, but now it's moved for expertise and population density. If you wanted to hire a 50,000 experienced assembly worked and 5,000 engineers tomorrow, China is the only place you could do it.

They do get medical and dental.

I'm not sure why this is being downvoted, its a valid comment.

Any stretched out supply chain for critical path items is at risk in any disruptive event. Globalization of your supply chain is fantastic at making stuff cheaper - but not at making the supply more durable.

Well the problem is that the critical equipment was not stockpiled... even if we could manufacture that we’d still have the problem of doing it quickly as the current ventilator shortage is showing. In fact one of the reasons things are made in China is not just that it’s cheap, but they have a large human workforce that can be adapted to quickly make or assemble all sorts of low complexity items.

> In fact one of the reasons things are made in China is not just that it’s cheap, but they have a large human workforce that can be adapted to quickly make or assemble all sorts of low complexity items.

This could be a chance to build new business models to help people retrain.

Nobody in the US wants to do these chinese jobs. I dont blame them

No one would think of/plan for a crisis that involves stocking 1m ventilators. Beyond that, that stock would need to be rotated out.

How can "RAID for supply chains" not just lead to being competed out of business by competitors without that added complexity in the short term? (The obvious answer of creating random disasters that hit single source competitors harder seems sketchy.) It seems like the only industries it work in would have to essentially be oligopoly class or otherwise noncompetitive to begin with.

This is key. Competition is about squeezing inefficiencies out. It is about accepting risk of supply chain disruption in order to reduce cost.

The more cynical way to put it is nobody gets a raise for a detailed continuity of operations plan. They do if they reduce manufacturing cost by 20%.

We have done some protectionism of other strategic industries, but those policies usually get a lot criticism when people aren't thinking that disaster is waiting around the corner.

See: farm subsidies

I think a lot of the criticism towards farm subsidies is less about the actual subsidy itself and more about perceived hypocrisy in the system for supporting only certain industries.

It doesn't help that many of those on the receiving end of the subsidies hold highly individualistic political positions while benefiting from diametrically opposed socialist policies.

China has an industrial policy, the United States prefers to leave things up to the free market.

https://prospect.org/special-report/industrial-policy-road-t...

This is changing, manufacturing is moving to Vietnam.

The reason why is because China is cheap. The CCP purposely deflated it's RMB so to attract foreign DTI. Direct trade investment. It also supported financing projects of infrastructure, factories and easy money lending in China in lieu for hiring chinese contractors to build in Chian and in turn then fuel the Capitalist West products that are made in China but "designed in USA" and then that knowledge is spread among other people in China in a copy/repeat system.

There's more, but it's mostly that, good education, focus on science, and he whole no religion thing and being a communist state meaning easy peasy to built factories, no right of way and no time consuming court system, no true intellectual property protection, and of course cheap labor.

With the emergence of good factories, also came good supply chains that were consistent within a single city so things can be deigned, built, tested and redone on scale, easily. Mini silicon valeys in a way.

Then transportation, highway, and then the Chinamail/EMS being subsidized by USPS to help offset delivery cost.

Vietnam is now the new star. :D

It distances us from the most local and immediate effects of pollution. China is ok with the pollution, we're not

> Why is everything made in China?

Because we don't let people migrate to the US in sufficient quantities to equalize the cost of labor. For better or worse, there's a ton of underemployed folks in China who's alternative to manufacturing is subsistence farming.

Since this is, in your own words, a national security problem, I imagine you're in favor of removing the national immigration quotas the US imposes?

In order for that argument to work, the US would also have to lower the minimum wage. Is that what you're proposing?

The strategy would presumably be halve the minimum wage while simultaneously halving the cost of everything else.

Someone one 60% of the US minimum wage in China would theoretically [0][1] have a more comfortable lifestyle than a US minimum wage worker. In all likelihood this is because they can buy stuff extremely cheaply because their manufacturing costs are lower.

[0] https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(PPP)...

[1] https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(nomi...

No need to, let enough people immigrate and the cost of wages in China will go up to over the US minimum wage

In your estimation, how many relocated people would be required to push the median Chinese wage over, say, $12?

Re 5.: At least the EU has suspended those rules.

https://www.euractiv.com/section/aviation/news/coronavirus-e...

The US suspended those rules on March 11:

https://thepointsguy.com/news/coronavirus-slot-restriction-s...

And planes are mainly being parked, not reused for cargo. Probably because of the capacity and fuel price issue.

Well, if the price for cargo delivery keeps going up I would expect that it would become profitable to un-park some of those planes.

Maybe. It sounds like the routes in demand are long haul flights, which I don't think a lot of narrow-body jets are capable of doing. Most jet airliners are narrow-bodies.

> Many companies already hedged the price of fuel so the oil glut isn't really helping them any.

It's really interesting that oil can be hedged against such a dramatic change. But not everything can be hedged in such a way. Do you have a simple explanation for what can and can't be hedged?

What matters for the financial hedge is the existence of futures, which means the companies can buy something they know they will need at a known price months before they actually need it. The specifics of this operation differs depending on the actual financial instrument used, but the idea is the same, even on the opposite side (selling at a known price in the future).

It is already a sunk cost if they just bought futures, so shouldn’t keep them from flying. They can take delivery to fly, or sell for someone else to take delivery at the low price.

The hedge must be more complicated to explain why they would stay grounded.

Hmm. I don't think the hedge is that complicated, so maybe it's not particularly relevant after all. I was just thinking of it in terms of complementary goods -- normally we'd expect suppliers to increase output when the cost of an input drops, but in this case they already ate that cost so the effect would be damped. But maybe not?

I suppose there's also a question of how long OPEC floods the market. But IDK, and you make a good point.

The futures contracts don't always guarantee delivery. "Cash-settled" instruments just pay out cash based on the market rate (with the expectation that the long party can purchase the oil at that price)

In this case they probably just signed a contract agreeing to pay $X/gallon for a period of time.

You can hedge any commodity like that if a seller exists to agree.

Except onions! :)

https://en.wikipedia.org/wiki/Onion_Futures_Act

And box office receipts.

https://en.wikipedia.org/wiki/Box_office_futures

Anything with a futures contract on major exchanges can be very easily. CME has a list of futures contracts: https://www.cmegroup.com/trading/products/

Anything can be hedged with a few legal exceptions) . That is what insurance is, a hedge against bad events. There are companies who specialize in insurance for one off events - for a price . For oil it works better because well drillers want to be sure they get enough money to pay off the well so someone is willing to take the other side and there is thus competition on both sides resulting in good deals.

As for 5. - this has been suspended for 4 months in Europe, so that airlines can stop running empty flights & still keep their slots:

https://www.euractiv.com/section/aviation/news/coronavirus-e...

And AliExpress sale?

Cargo and passengers typically cross subsidise each other, depending on the route.

Without (business and first class in particular) passengers the costs of running the planes have to met fully by the limited cargo.

I wonder how much this is a shortage of supply vs. an increase in demand. From what I understand, commercial passenger flights often take commercial cargo. With fewer flights, all that demand for cargo capacity has to go somewhere.

There is only so much air cargo capacity, and passenger aircraft cannot be rapidly converted into that capacity. The cost is excessive to convert passenger aircraft to cargo service.

Yea that's my assumption. Don't know what it takes to pull out seating, overheard, flooring?, etc. and what sort of effort that takes time wise. I assume it's not simple and takes time.

The doors are (one of) the problems.

For smallish cargoes I don't see why you'd need to rip out anything. Place it on the floorspace in front of seats, place it on the seats. Getting it in an out is a matter of people lifting and carrying.

(am not an air cargo freight expert if you haden't realised already, so be gentle)

I'd imagine that if these parcels are not strapped down and the load shifts in flight because of severe turbulence, that could shift the center of gravity enough to result in a chain of events leading to a crash.

You have to secure the loads or they can shift, causing the airplane to crash. I agree that you take out the seats in the aircraft but I don't know if the cargo "hooks" are certified or even there. Cargo can not be loaded into aircraft if it can not be secured into place.

Seats are installed via a track system present on the floor. Seats are part of what’s called BFE, and a flexible track system means you’re not drilling holes in the floor.

Limited loads could be secured in the passenger cabin using the same tracks. Whether or not that actually makes economic sense is a different discussion.

You can see this in action here: https://www.reddit.com/r/aviation/comments/8scjkk/ever_wonde...

Looks like someone could start making cargo containers that secure into BFE, once the seats are removed...

I seriously thought that the need to strap down or generally secure this stuff, and that it would be done, was so obvious as to be beneath mentioning. It seems not.

How do you properly secure cargo in receptacles intended to secure humans?

German post uses it for next-day letter delivery. They just put the boxes in bags and use the seatbelts to stap the bags to the seats.

Is that really an unsolved problem in airline business? It can't be that hard to build small containers that fit on the seat and can be fastened with the seat belt system.

Obviously, you won't want to do this for general cargo transport, but for "hey, we have a grounded plane, but transporting air mail gives great rates right now" situations?

So it's not a solved problem on massive scale because there aren't frequent and long worldwide air passenger traffic shutdowns.

Don’t passenger planes already have huge cargo holds for all the luggage? Would it really be that hard to stuff that space full of cargo?

I would guess typical passenger planes have significantly smaller cargo stores than cargo planes so the layout may not be good for efficient cargo transport (outside of course... people as cargo).

You'd allocate the volume in the aircraft very differently I imagine for cargo than passengers. I also suspect passenger planes were already using any leftover cargo from passenger cargo storage for commercial cargo transport to begin with. Two potentially very different optimization problems.

When airlines are flying people across the country for $13, it makes all of those look like second or third order effects:

https://news.ycombinator.com/item?id=22731785

It's not hard, just more expensive. I imagine not having 200-some people paying, let's say $150, would mean air cargo would have to cover the $30,000-ish difference. There's additional cargo space because the baggage hold is no longer taken up by passenger baggage, but with airlines charging passengers $50/bag, we might be able to guess at the opportunity cost to airlines for taking something the average size and weight of a passenger's checked bag.

> It's not hard, just more expensive. I imagine not having 200-some people paying, let's say $150, would mean air cargo would have to cover the $30,000-ish difference.

I think you are drastically underestimating the typical revenue of a long haul or international flight.

A good starting point is RASM, or average revenue per available seat mile. For the airline industry a reasonable benchmark for that is about 15 cents, or so.

So multiply the number of seats on the plane by the number of miles and you'll get something that's at least in the same order of magnitude for passenger revenue on a flight.

Since we know air freight only really makes sense for long distances or over water -- you're going to be using a truck to get medical supplies from Dallas to NYC, and it will take less than 24 hours -- let's pick a good example route:

HKG > LAX on a 767, a standard wide-body

7,243 miles x 375 seats x $0.15 = $406,912.50

Not the person you were replying to, but that is definitely much higher than I would have guessed. How do you derive the $0.15 / a mile you used as a rough estimate?

I mean it just kind of has to be higher for anything to make sense. The depreciation/fixed costs of operating a 767 for an hour is around $10k plus another $20k or so for fuel. That's a couple hundred grand for an 8 hour flight before you even get started with the expense of actually running an airline as a business.

So we know the post guessing $30k in passenger revenue for a typical flight is in the wrong order of magnitude.

For the number I kind of picked 15 cents out of my ass, it's a number I've seen in the past year in earnings reports for the majors and it easily gets us into the basic range we're talking about. Just google the terms RASM, PRASM, and CASM and you'll find a ton of info, that's the basic jargon used in the business to come up with info of this type.

Thank you for the reply. Knowing which keywords apply helps. Much appreciated!

Yup, they're already doing that: https://www.forbes.com/sites/jamesasquith/2020/03/28/commerc...

Passenger planes already take cargo as part of normal flights yes!

Now there are less passenger flights, so less capacity there! And less passengers to make the flight break even with people + cargo paying.

Not comparable to dedicated cargo aircraft.

Why? I'm genuinely curious. Obviously total cargo volume is lower. Is there some regulatory concern that makes them "not comparable"?

The volume vs weight probably comes into play as well.

Without passengers, passenger aircraft probably run out of volume capacity long before they approach their max take off weights.

Whereas I think with dedicated cargo aircraft, it is usually the other way around - weight is the main limit, less volume.

I'm guessing they do a lot more pallet shipping, and your typical passenger jet isn't prepped to load / store / unload that versus luggage.

The costs are actually reflecting the repurposing of passenger planes for cargo flights, at least on a small scale. Most passenger planes have cargo space...it's about 1/2 to 1/3 the usable space of their plane, therefore they are quite a bit less efficient. The increase in prices is reflective of a new market clearing price where capacity can meaningfully be increased.